Human Molecular Genetics, 2008, Vol. 17, No. 17 2595–2609 doi:10.1093/hmg/ddn159 Advance Access published on May 23, 2008 Sex differences in a transgenic rat model of Huntington’s disease: decreased 17b-estradiol levels correlate with reduced numbers of DARPP321 neurons in males

1,{ 1,{ 1 1 2

Felix J. Bode , Michael Stephan , Hendrik Suhling , Reinhard Pabst , Rainer H. Straub , Downloaded from https://academic.oup.com/hmg/article-abstract/17/17/2595/632381 by guest on 27 June 2020 Kerstin A. Raber3, Michael Bonin4, Huu Phuc Nguyen4, Olaf Riess4, Andreas Bauer5,6, Charlotte Sjoberg7,A˚ sa Peterse´n7 and Stephan von Ho¨ rsten1,3,

1Institute of Functional and Applied Anatomy, Medical School of Hannover, 30625 Hannover, Germany, 2Department of Internal Medicine I, University Regensburg, 93042 Regensburg, Germany, 3Experimental Therapy, Franz-Penzoldt- Center, Friedrich-Alexander-University Erlangen-Nu¨rnberg, 91054 Erlangen, Germany, 4Department of Medical Genetics, University of Tu¨bingen, 72076 Tu¨bingen, Germany, 5Institute of Neurosciences and Biophysics, Research Centre Juelich, 52425 Juelich, Germany, 6Department of Neurology, Heinrich-Heine University Duesseldorf, 40001 Duesseldorf, Germany and 7Translational Neuroendocrine Research Unit, Department of Experimental Medical Science, Lund University, 22184 Lund, Sweden

Received March 10, 2008; Revised May 8, 2008; Accepted May 21, 2008

Recent clinical studies have highlighted that female sex hormones represent potential neuroprotective mediators against damage caused by acute and chronic brain diseases. This evidence has been confirmed by experimental studies documenting the protective role of female sex hormones both in vitro and in vivo, although these studies did not specifically focus on Huntington’s disease (HD). We therefore investigated the onset and course of HD in female and male transgenic (tg) HD (CAGn51) and control rats across age and focused on three aspects: (i) behavioral and physiological alterations (energy expenditure, home-cage activity, emotional disturbance and motor dysfunction), (ii) morphological markers (numbers and characteristics of striatal DARPP321 medium-sized spiny neurons (MSNs) and dopamine receptor autoradiography) and (iii) per- ipheral sex hormone levels as well as striatal estrogen receptor expression. Independent of their sex, tgHD rats exhibited increased levels of food intake, elevated home-cage activity scores and anxiolytic-like behavior, whereas only males showed an impairment of motor function. In line with the latter finding, loss and atrophy of DARPP321 MSNs were apparent only in male tgHD rats. This result was associated with a decreased striatal dopamine D1 receptor density and lower plasma levels of 17b-estradiol at the age of 14 months. As DARPP321 MSNs expressed both a-andb-estrogen receptors and showed a correlation between cell numbers and 17b-estradiol levels, our findings suggest sex-related differences in the HD phenotype pointing to a substantial neuroprotective effect of sex hormones and opening new perspectives on the therapy of HD.

INTRODUCTION striatal neurodegeneration (2). The clinical triad of motor dys- function, cognitive decline and psychiatric manifestations Huntington’s disease (HD) is a polyglutamine disorder based on characterizes HD. Owing to its autosomal-dominant inheri- an expanded CAG triplet repeat (1) leading to cerebral and tance, the prevalence is equally distributed between both

To whom correspondence should be addressed at: Experimental Therapy, Franz-Penzoldt-Center, Friedrich-Alexander-University Erlangen-Nu¨rnberg, Palmsanlage 5, 91054 Erlangen, Germany. Tel: þ49 91318523504; Fax: þ49 91318523502; Email: [email protected] †The authors wish it to be known that, in their opinion, the first two authors should be regarded as joint First Authors.

# The Author 2008. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected] 2596 Human Molecular Genetics, 2008, Vol. 17, No. 17 sexes (3,4). However, potential sex differences concerning the life, there was a parallel weight gain in male tgHD rats and age of onset and the course of the disease are poorly defined, controls. In later life, male tgHD rats showed a trend as difficulties of matching female and male HD patients regard- towards lower body weights (Fig. 1), in line with previous ing their CAG repeat lengths limit comparability. There are reports (13). In contrast, two factorial analysis of variance only few reports focusing on this issue, suggesting that the (ANOVA) for repeated measurements revealed a significant age of onset of HD is higher (5) and the course of disease is effect for the factor genotype, as female tgHD rats gained more moderate (6) in women compared with men. Analogously, more weight than controls (F1,270 ¼ 4.3, P , 0.05). This was sexual dimorphisms regarding the prevalence and severity of mainly due to an increased weight gain in the first 4 months disease have also been reported for other neurodegenerative of life (Fig. 1). diseases, e.g. Parkinson’s and Alzheimer’s disease (7). To investigate whether this altered weight gain was associ- Thus, although clinical studies provide clear evidence that ated with differences in feeding behavior, we determined the female sex hormones exert neuroprotective effects in acute individual daily food intake. Although ANOVA for the Downloaded from https://academic.oup.com/hmg/article-abstract/17/17/2595/632381 by guest on 27 June 2020 and chronic brain diseases, this has not been unequivocally factor genotype showed a significantly increased absolute documented in HD (8,9). Likewise, experimental studies doc- food consumption (F1,216 ¼ 12.1, P , 0.01) as well as relative umenting a protective role of female sex hormones both in food intake in female tgHD rats as compared with controls vitro and in vivo have so far not specifically focused on HD (Fig. 1C and E; F1,216 ¼ 7.9, P , 0.05), the relative food (10,11). In order to examine sex-related differences in HD, intake (%/body weight) was significantly elevated only in animal models provide a useful approach to study sex differ- tgHD males compared with wild types (Fig. 1D and F; ences, because they allow the circumventing of the problems F1,216 ¼ 14.1, P , 0.01). Thus, male rats ingested more food of standardization, which are regularly occurring in humans. absolutely (P , 0.0001), but ate significantly less compared We therefore investigated the phenotype of male and female with females in relation to their body weight (P , 0.0001). transgenic (tg) HD rats carrying 51 CAG repeats under the Analyzing the indirect calorimetry using a dedicated calori- control of the endogenous rat huntingtin promoter (12). In metric system (TSE LabMaster system) revealed significant this HD model, male animals exhibit a progressive neuro- sex differences (ANOVA for repeated measurements: pathological phenotype closely resembling the most common F1,2280 ¼ 28.4, P , 0.0001) with female tgHD rats showing late manifesting type in human HD (13). However, a few a trend towards increased calorimetric exchange rate (kcal/h/ studies using only female rats have suggested a blunted pheno- kg) (F1,1216 ¼ 3.9, P ¼ 0.06, n.s.) (figures illustrating VO2, type in female tgHD (14–16), but this was not directly com- VCO2 and consecutive calorimetric measures are available pared with an additional male cohort. online). We therefore aimed to characterize the behavioral phenotype The diurnal baseline activity in the home-cage environment of female tgHD compared with their male counterparts as well was monitored. Here, male rats tg for HD showed an elevated as control littermates across age. Here, the focus was on activity compared with controls. This was detectable at all energy expenditure, activity in the home-cage, emotional dis- time points investigated, whereas females did not show turbance within the social interaction test as well as motor dys- significant hyperactivity until the age of 4 months. In later function using the accelerod test. As female tgHD rats showed life, tgHD females showed a dramatically increased most of the behavioral abnormalities seen in males, the most hyperactivity, whereas female wild-types and both tgHD and prominent difference was a lack of motor impairments in wild-type males remained on moderate activity levels females. As the striatum plays a key role in the coordination (Fig. 2). As exemplified in Figure 2 for the different ages, of motor function and is one of the key brain areas affected in ANOVA for repeated measurements with the interindividual HD (2), we focused on possible morphological alterations in factor genotype and the intraindividual factor ‘activity over the striatum by means of stereology of medium-sized spiny time’ revealed significant effects for young rats (females: neurons (MSNs) expressing dopamine- and cAMP-regulated F1,1566 ¼ 1.6, P , 0.001; males: F1,1566 ¼ 1.7, P , 0.0001) þ phosphoprotein of 32 kDa (DARPP32 ), dopamine receptor as well as for 14-month-old rats (females: F1,1566 ¼ 7.5, expression and concomitant changes in sex hormone plasma P , 0.0001; males: F1,1566 ¼ 1.5, P , 0.01). To illustrate levels. Our findings and conclusions were compared with a this hyperactivity, videotapes highlighting genotype- re-analysis of publicly available -expression profiling dependent effects on the home-cage activity in tgHD rats data from HD patients and controls (17), which were split by compared with wild-types at the age of 8 months are provided sex and discussed with regard to estrogen receptors (ER), with the Supplementary Material. Similar results were aromatase (CYP-19), , steroid-5-alpha-reductase obtained scoring beam break-based activity in the LabMaster (SRDA), peroxisome proliferator-activated receptor gamma system (data not shown). co-activator alpha (PGC-1a), B-cell lymphoma-associated The classical HD triad also includes emotional disturbance athanogene 1 (BAG1), gamma-aminobutyric acid (GABA) and motor dysfunction, which can be tested repeatedly across and insulin-like growth factor receptor 1 (IGFR1) expression. lifetime. To further monitor the progress of the disease, anxiety-like behavior was measured using the social inter- action test of anxiety as well as motor coordination and RESULTS balance in the accelerod test. In both male (F1,108 ¼ 16.1, P , 0.01) and female (F1,108 ¼ 24.9, P , 0.0001) tgHD rats, Blunted behavioral phenotype in tgHD females there was a significant increase of the parameter ‘time spent As a loss of weight is often seen in HD, body weight was con- in active social interaction’ (revealed by ANOVA for repeated tinuously measured in all rats across age. In the first months of measurements, factor genotype). This indicates emotional Human Molecular Genetics, 2008, Vol. 17, No. 17 2597 Downloaded from https://academic.oup.com/hmg/article-abstract/17/17/2595/632381 by guest on 27 June 2020

Figure 1. Body weight and food consumption across age. The left column indicates data obtained from females and the right column indicates data from males. (A and B) The graphs illustrate the weight gain for tgHD rats (filled squares) as well as for wild types (open circles) from the age of 4 to 58 weeks. (C and D) Food intake was measured once a month over 3 days for each rat and the mean absolute intake per day was calculated. (E and F) This daily food intake was related to the body weight and illustrated as food intake per 100 g body weight. The inserts show the means of each graph over the whole period. Asterisks indicate significant differences between means (bar graph) of the respective measurements (n ¼ 10/group).

disturbances in both sexes (Fig. 3A and B). However, when Morphological impairment in tgHD males only investigating onset and progress of motor dysfunction, only male tgHD rats showed a significant impairment compared To explore whether the phenotype of tgHD males is associated with controls (F1,96 ¼ 4.8, P , 0.05). Remarkably, no motor with neuropathological changes in the striatum, stereology- dysfunction was detectable by this well-established test in based determination of absolute numbers of MSNs as well female tgHD rats (Fig. 3C and D). as striatal dopamine receptor autoradiography was performed. 2598 Human Molecular Genetics, 2008, Vol. 17, No. 17 Downloaded from https://academic.oup.com/hmg/article-abstract/17/17/2595/632381 by guest on 27 June 2020

Figure 2. Home-cage hyperactivity is most pronounced in tgHD females. Baseline activity was measured in home-cages with infrared-sensors over a period of 22 h every month. Graphs illustrate basal activity of 2-, 8-, and 14-month-old rats in intervals of 15 min for females on the left and males on the right side. For a rapid overview, the mean activity over the 22 h measured is inserted into each graph as an additional bar chart. Activity of transgenic rats is indicated in black compared with wild-types (white). Development of the duration of large movements (.1 s) is indicated as large time (LT) in seconds per 15 min, with the grey background showing the 12 h of the dark cycle. While there is a stable but slight hyperactivity in males over the whole life span (B: 2nd month; D: 8th month; F: 14th month), there are no differences in females at the age of 2 months (A). However, there is a significant increase of daily activity in the older females (C: 8th month; E: 14th month). Asterisks indicate significant differences between means (bar graphs) of the respective measurements (n ¼ 10/group).

Stereological analysis provided three different parameters: was detected in the tgHD males compared with control litter- (a) striatal volumes, (b) total number of DARPP32þ cells in mates at the age of 14 months (Fig. 4D; P , 0.01). Addition- the striatum and (c) cross-sectional cell body areas of ally, two-factor ANOVA of the data from 4- and 14-month-old DARPP32þ neurons. Representative immunohistological rats using genotype and sex as factors revealed significant þ staining of DARPP32 MSNs is illustrated in Figure 4. interactions between sex and genotype (4 months: F1,5874 ¼ In general, the striatal volume increased across groups from 43.2, P , 0.0001; 14 months: F1,4681 ¼ 53.9, P , 0.0001) þ 4 to 14 months (F1,26 ¼ 48.1, P , 0.0001), but there were no counting DARPP-32 cells. Further analysis showed a signifi- changes detectable either between females and males or cantly higher average of the cross-sectional areas only in the between tgHD and wild-type animals (Table 1). Furthermore, 4-month-old tgHD males (P , 0.0001). A significantly no differences of the total cell number at the age of 4 months smaller average was found in the 14-month-old tgHD males were found (Fig. 4C). A two-factor ANOVA of the data from (P , 0.0001), whereas the tgHD females revealed a signifi- 14-month-old rats revealed a significant interaction between cantly larger cell size at this age (P , 0.0001). the factors sex and genotype (F1,12 ¼ 11.6, P , 0.01). As most of the MSNs express dopaminergic receptors (18) Although no differences in the striatal volumes either and dopamine receptors are altered in the striatum (19), D1 between females and males or between tgHD and wild-type and D2 receptor densities in the striatum were examined animals were found, a significant loss of DARPP32þ cells using autoradiography. Here, only in tgHD males a significant Human Molecular Genetics, 2008, Vol. 17, No. 17 2599 Downloaded from https://academic.oup.com/hmg/article-abstract/17/17/2595/632381 by guest on 27 June 2020

Figure 3. Impaired motor function in tgHD males only. Anxiety-like behaviors (reduced activity in the social interaction test of anxiety) as well as motor coordination and balance (accelerod) were measured every other month starting at the age of 2 months. Both tgHD sexes (black squares) show a significant increase in social interaction time compared with their controls (open circles; n ¼ 10/group; A, B). While the accelerod test revealed no changes in the females (n ¼ 10/group; C), tgHD males perform significantly worse than their control males (n ¼ 9/group; D). Inserts show the average performance of the whole test period. The best performance of two trials on each test day is given, measured in rotations per minute (rpm). Asterisks indicate significant differences between means (bar graphs) of the respective measurements.

reduction of the striatal D1 receptor density across 4 to 14 months (F1,18 ¼ 4.8, P , 0.05) turning into significantly months of age was found (Fig. 5A and B; F1,8 ¼ 8.2, decreased levels compared with controls at the age of 14 P , 0.05), whereas neither tgHD female rats nor controls months (Fig. 6B; F1,17 ¼ 11.1, P , 0.01). Further analysis of showed any changes (Fig. 5C–H). Analysis of D2 receptor 17b-estradiol revealed a significant reduction in tgHD males density revealed no significant differences independent of from 4 to 14 months (F1,17 ¼ 4.8, P , 0.05) as well as the sex, age and genotype (data not shown). reduced levels of 17b-estradiol comparing 14-month-old tgHD males with females (F1,16 ¼ 9.9, P , 0.01). DHEA levels increased in all four groups over time (Fig. 6C b Reduction of 17 -estradiol levels in symptomatic tgHD and D), whereas significant genotype effects with higher levels males in the tgHD rats occurred only in females (Fig. 6C; at 4 As sex-specific differences in the HD phenotype and its neuro- months: F1,30 ¼ 14.5, P , 0.01; at 14 months: F1,31 ¼ 7.4, pathology were found, plasma levels of 17b-estradiol, dehy- P , 0.05). In males, no differences for DHEA levels were droepiandrosterone (DHEA), testosterone and estrone were observed (Fig. 6D). analyzed. As expected, ANOVA revealed significantly Testosterone levels were significantly elevated in increased levels of 17b-estradiol (P , 0.01) and estrone 4-month-old tgHD females compared with controls (F1,16 ¼ (P , 0.0001) in females compared with males as well as 26.2; P , 0.0001) and also significantly higher in increased levels of DHEA (P , 0.0001) and testosterone 14-month-old control females compared with 4-month-old (P , 0.0001) in males compared with females (Fig. 6). controls (Fig. 6E; F1,18 ¼ 18.4, P , 0.01). TgHD males Comparing tg and wild-type rats of the same sex revealed showed only a trend towards lower levels in 14-month-old that female tgHD rats showed a trend towards increased rats compared with 4-month-old ones (Fig. 6F). 17b-estradiol levels (Fig. 6A). tgHD males displayed signifi- For estrone, we found significantly higher levels in females cantly increased levels of 17b-estradiol at the age of 4 compared with males in all groups except for 4-month-old 2600 Human Molecular Genetics, 2008, Vol. 17, No. 17 Downloaded from https://academic.oup.com/hmg/article-abstract/17/17/2595/632381 by guest on 27 June 2020

Figure 4. Reduced numbers and atrophy of striatal DARPP32þ MSNs in symptomatic tgHD males. DARPP32þ staining shows a specific labeling of ~95% of striatal neurons (A, overview, scale bar 1000 mm; B, higher magnification, scale bar 50 mm). Black bars show tgHD and white bars show wild-type animals (n ¼ 4–5/group). At the age of 4 months, no differences concerning the total number of DARPP32þ neurons are apparent (C), while at the age of 14 months tgHD male rats show a significant reduction in DARPP32þ neurons (D). The cross-sectional area of cell bodies was measured using the nucleator prin- ciple (.200 neurons/brain evaluated). Male tgHD animals show significantly larger areas compared with their controls (E) at the age of 4 months, which changes to significantly smaller areas at the age of 14 months (F). Females show a difference only at the age of 14 months with increased areas in the tgHD. tgHD animals. Furthermore, estrone levels decreased from modification of MSNs DARPP32þ signaling by 17b-estradiol 4- to 14-month-old tgHD females (data not shown). occurs via ERs. Therefore, screening for co-expression was As sex differences were observed, it was investigated carried out, allowing such an interaction using double immu- whether the behavioral and/or morphological impairments nofluorescence of ER-a and ER-b and DARPP32þ. It has correlated with sex hormone levels. Regression analyses been reported that both types of ERs are expressed in the stria- revealed a significant correlation of 17b-estradiol with the tum of mice (20), but their cellular localization is obscure. As total number of DARPP32þ MSNs (Fig. 6G) as well as with a significant decrease of 17b-estradiol in tgHD males of 14 cell numbers/mm3 (Fig. 6H). Further significant correlations months was found and as 17b-estradiol is thought to exert neu- were found for 17b-estradiol (P , 0.05), estrone (P , 0.05), roprotective effects (21), it was hypothesized that ERs are also DHEA (P , 0.01) and testosterone (P , 0.0001) with body expressed on MSNs. Therefore, a three-color immunofluores- weight, due to set effects when studying sex differences. cent staining was performed for nuclei, DARPP32þ (as a The correlation of 17b-estradiol with the numbers of marker for MSNs), and either ER-a or ER-b. We found that DARPP32þ MSNs led us to hypothesize that a direct ERs-a were expressed on DARPP32þ neurons in the striatum Human Molecular Genetics, 2008, Vol. 17, No. 17 2601

Table 1. Striatal volumes

Females Males 2/2 wt þ/þ tg 2/2 wt þ/þ tg

4 months 31.3 + 0.5 32.1 + 0.8 33.5 + 0.8 32.0 + 0.6 14 months 35.6 + 0.6 35.1 + 0.8 36.1 + 0.4 37.0 + 1.4

Average striatal volumes assessed unilaterally (mean mm3 + SEM) using calculated repeated measures of striatal areas across the striatum (11.28 mm Interaural/2.28 mm Bregma and 8.28 mm Interaural/20.72 mm Bregma) in 4- and 14-month-old female and male ¼ rats. Wild-type (2/2), transgenic rats (þ/þ), each group n 4–5. Downloaded from https://academic.oup.com/hmg/article-abstract/17/17/2595/632381 by guest on 27 June 2020 P , 0.05, 4 versus 14 months.

(Fig. 7A–D). Furthermore, there was a clear co-expression of ERs-b on DARPP32þ neurons (Fig. 7E–H). Signals for both ERs were detectable in females and males as well as in tgHD and controls rats, showing no obvious differences of expression in any group.

Gene-expression profiling of human HD striata reveals downregulation of GABA-associated In order to link our findings to the human condition, publicly available gene expression profiling data (17) were split by sex and re-analyzed. Analysis of 164 ‘hypothesis-based’ genes associated with the steroid system revealed a limited number of transcripts that were significantly differentially regulated. When comparing female controls with male controls (Sup- plementary Material, Table S1), the GABAA2 receptor (GABRA2) was found to be significantly downregulated in females, whereas the GABAA5 receptor (GABRA5), ER alpha (ESRRA), CYP51A1 and steroid sulfatase S (STS) Figure 5. Selective loss of D1 receptors in the striatum of tgHD males. The density of dopamine receptors in the striatum was assessed by receptor auto- were upregulated. The comparison of HD females with HD radiography. Representative autoradiographs of coronal sections of 4 month males (Supplementary Material, Table S2) revealed no signifi- (left side) and 14 months (right side) old rats are given (A–H). Significant cantly regulated transcripts. Comparing HD females with reductions in receptor densities are found in tgHD between the 4- and control females (Supplementary Material, Table S3) as well 14-month-old groups (A and B), while no differences are detectable in the as the comparison of HD males with control males (Sup- other groups (C–H). Color scale at the bottom illustrates the values of absol- ute receptor density (n ¼ 4–5/group). plementary Material, Table S4) revealed several significantly downregulated transcripts of GABAA receptor subunits with GABRA5 being most prominent. Only in male HD patients, transcripts for insulin-like growth factor receptor 1 (IGFR1) the -associated enzyme steroid-5-alpha-reductase is potentially attributable to a differential estrogen signaling. (SRD5A1) was significantly downregulated and in none of Furthermore, results showed neither significant difference in the HD patient groups a differential expression of transcripts CYP-19 nor in PGC1 or BAG1 transcripts (Supplementary coding for ERs was found. Material, Tables S1–S8). Analysis of transcripts with a significant 1.5-fold change and focusing on potential sex differences revealed in the com- parison of female controls with male controls (Supplementary DISCUSSION Material, Table S5) 343 differentially regulated transcripts including sex- linked genes. The comparison of This is the first report to comprehensively characterize the HD females with control females (Supplementary Material, behavioral, physiological and neuropathological profiles in Table S6) revealed 1770 with eight GABA-associated genes both sexes of an animal model of HD across age. Although and the comparison of HD males with control males (Sup- alterations of energy expenditure, home-cage activity scores plementary Material, Table S7) 2520 with 15 and emotionality were obvious in both sexes, motor dysfunc- GABA-associated significantly altered transcripts. Interest- tion in the accelerod test was only observed in tgHD males ingly, a very pronounced downregulation of the dopamine when monitored across the first 14 months of life. This receptor D1 (DRD1) was evident in male HD patients. In con- finding was associated with a loss and atrophy of trast to the other comparisons, the comparison of HD females DARPP32þ MSNs as well as a lower density of the striatal with HD males revealed only 42 differentially expressed tran- dopamine D1 receptors in symptomatic male but not in scripts. Out of those, only a significant downregulation of female tgHD rats. As we also report lower plasma levels of 2602 Human Molecular Genetics, 2008, Vol. 17, No. 17 Downloaded from https://academic.oup.com/hmg/article-abstract/17/17/2595/632381 by guest on 27 June 2020

Figure 6. Reduced estradiol levels in symptomatic tgHD males only. Graphs illustrate plasma levels of sex hormones in tgHD (black columns) and correspond- ing wild-types (white columns) with females on the left and males on the right side. Only tgHD males show significant changes in 17b-estradiol with increased levels in 4-month-old and significantly decreased levels in 14-month-old tgHD males (B), while no significant differences are apparent in females (A). DHEA levels increase in all four groups from 4 to 14 months, while only tgHD females show significantly higher levels in comparison with controls. Testosterone levels for female (E) and male (F) rats show a clear sex difference as well as elevated testosterone levels in 4-month-old tgHD females compared with their controls. Regression analysis of 17b-estradiol levels with total number of DARPP32þ cells (G; y ¼ 2150380.3 þ 22297.4x; R2 ¼ 0.3, P ¼ 0.02) as well as with the number of cells per mm3 (H; y ¼ 25605.1 þ 686.9x; R2 ¼ 0.3, P ¼ 0.02) reveals significant correlations. Asterisks indicate significant differences (n ¼ 8–10/group). Human Molecular Genetics, 2008, Vol. 17, No. 17 2603 Downloaded from https://academic.oup.com/hmg/article-abstract/17/17/2595/632381 by guest on 27 June 2020

Figure 7. Expression of both ER-a and -b on DARPP32þ neurons in the striatum. Micrographs provide evidence of a co-localization of DARPP32þ neurons with ERs-a (A–D) and with ERs-b (E–H). Nuclear staining is shown with DAPI (blue, A and E). While the ER-a-expression (B, green) is not strong, the expression of ER-b (F, green) is more distinct on DARPP32þ neurons (C and G, red). Merged images are displayed on the right-hand side of each row (D and H). Arrows highlight double-stained neurons.

17b-estradiol correlating with decreased numbers of striatal emotional disturbance might at least partly be due to an neurons in these symptomatic male tgHD rats, our finding of altered stress response, since progressive alterations in the DARPP32þ MSNs expressing both a- and b-ERs provides a hypothalamic–pituitary–adrenal axis activity have recently potential mechanism underlying the dramatic sex differences been reported for R6/2 mice (28). For these mice and, even in this model of HD. Our data point to a neuroprotective more importantly, also for humans suffering from HD, evi- action of sex hormones, which might provide a target for neuro- dence of hypothalamic changes is emerging including protective therapy aiming at postponing the onset and reducing altered levels of emotion-regulating neuropeptides expressed the severity of HD. in this brain region (29). Loss of body weight is a common finding in HD patients. Furthermore, our results extend previous reports showing Mochel et al. (22) focused on this aspect in presymptomatic, sex differences in behavioral read-outs in a neurotoxic rat early and mildly symptomatic HD patients and found a signifi- model (30) and a knock-in mouse model of HD (31). Giordano cant weight loss in both sexes. However, a significantly and Mejia-Viggiano (30) focused on locomotor activity, reduced body mass index occurred only in HD men despite a revealing an increase in both sexes with lesions compared higher calorific intake. An increased metabolism with a with untreated controls and in females compared with males. higher calorific intake has recently also been shown in young In tgHD rats, similar effects occurred in the home-cage test. male mice of the R6/2 model. However, data on female mice Interestingly, Dorner et al. (31) also found increased are still lacking (23). In the present study, we found a trend activity in females in the knock-in HD mouse model compared to lower body weights in the tgHD male rats, thereby confirm- with males using a home-cage based running wheel activity ing our previous report of a significant reduction of weight gain within the dark cycle, whereas in contrast, the total activity in symptomatic tgHD rats (13). In partial contrast, tgHD time in the open-field test was higher in males. Surprisingly, females showed an increase of body weight compared with they found no motor dysfunction in the rotarod test at all, wild types over their whole lifetime. Owing to a reduced although it is a common finding in several models of HD lean body mass despite a higher calorific intake in HD men (32,33). The tgHD rats showed impaired motor function only, Mochel et al. (22) postulated a hypermetabolic state in when tested in the accelerod test (13,25). This impairment early stages of HD. In our model, both tgHD sexes revealed seems to occur only in male tgHD rats, as we found no deterio- a higher relative caloric intake compared with wild-types. ration in tgHD females up to 14 months of age. In humans, a Owing to the loss of body weight only in males, our data later onset and a moderate time course have been suggested also suggest a hypermetabolic state in male tgHD rats. There- for female patients suffering from HD (5,6). Therefore, it is fore, the model of tgHD rats represents a useful tool that cor- possible that a later onset of classical motor dysfunction responds to the metabolic changes observed in HD patients. might also occur in female tgHD rats at even later time-points. The classical HD triad of symptoms is composed of cogni- As the most prominent finding concerning sex differences tive decline, emotional disturbance and motor dysfunction was the lack of motor dysfunction in females, we focused (24). Using the parameter ‘time spent in active social inter- on sex-dependent morphological alterations in the striatum. action’ in the social interaction test of anxiety, our study pro- We found a loss of DARPP32þ neurons in male tgHD rats, vides evidence of emotional affection in both sexes of tgHD compared with wild-type littermates, age-matched female rats, a finding that confirms previous results in male tgHD tgHD rats as well as to young tgHD males. These rats (25) as well as female R6/2 mice (26), which might be DARPP32þ neurons represent more than 95% of the MSNs attributed to pathological processes affecting the ventral stria- in the striatum (34). Losses of these cells have previously tum and the extended amygdala (27). In addition, this been reported for YAC128 mice (32), but for R6/2 mice 2604 Human Molecular Genetics, 2008, Vol. 17, No. 17 there are partly contradictory reports (35,36). For the rat model recognition sites with intracellular receptors mediating of HD, the loss of MSNs exclusively in male tgHD represents delayed actions with a prolonged duration, whereas especially a novel finding. It exceeds by far the total striatal neuronal cell neurosteroids bind to membrane recognition sites inducing fast loss reported previously in several months older female tgHD effects (49,50). Here, we report that peripheral sex hormone rats (11). Furthermore, we observed an increased cell size in levels correspond to a more severe clinical course in tgHD young tgHD males that turned into a decreased cell size in males. This points to 17b-estradiol being a key neuroprotec- older ones. As there was also an increase of cell size in tive factor in female tgHD rats, as this hormone might exert female tgHD rats at the age of 14 months and no neuronal its neuroprotective action via at least three different and inde- cell loss, this might indicate a delayed onset and progress of pendent pathways (9): (1) genomic, (2) nongenomic (rapid) the HD-related pathobiology in females, being suggestive of and (3) antioxidant, with the latter only being observed at endogenous neuroprotective mechanism in females. The pharmacological concentrations. For both the genomic and enlarged cell size in young male and old female tgHD rats the non-genomic pathway, ERs have been described as essen- Downloaded from https://academic.oup.com/hmg/article-abstract/17/17/2595/632381 by guest on 27 June 2020 showing no striatal loss might indicate compensatory mechan- tial (9,21), which prompted us to study the co-expression of isms and correspond to changes in cortex morphology of pre- both ERs (a and b) on DARPP32þ MSNs in the tgHD rat clinical patients, exhibiting enlargement of gyral crowns and striatum. As we found direct evidence of both receptors thin sulci (37). being expressed on these neurons this pattern provides the ana- The regulation of the state of DARPP32þ phosphorylation tomical basis for long-lasting effects of 17b-estradiol on at Thr34 provides a mechanism for integrating information MSNs. Recently, a neuroprotective action of 17b-estradiol (38). This phosphorylation is mediated via the D1, with the was reported on DARPP32þ neurons in vitro (11), providing D2 receptor counteracting this process (38–41). Therefore, further evidence of the mechanisms proposed here and point- we investigated the density of dopamine receptors in the stria- ing towards an important role of sex hormones in neurodegen- tum using receptor autoradiography. There were no changes in erative disorders, particularly in those affecting the striatum D2 receptor density in tgHD rats until the age of 14 months, (51). although this has been reported for human beings (19) and In addition, centrally produced estrogens might also have older tgHD rats (35). However, D1 receptor density initially influence on proliferation and survival of neurons either increased at 4 months and decreased at 14 months of age. directly or via interaction with insulin-like growth factor-I This apparently dynamic upregulation followed by significant (IGF-I) (52). Furthermore, neurons in several brain regions downregulation in young versus old tgHD male rats is in line express aromatase (the CYP19 gene product) that converts with previous findings showing that very old male tgHD rats neuronal- and glial-derived testosterone into 17b-estradiol exhibit a dramatic loss of D1 receptor density at about 24 (for review, see 53–55). Apart from the association of months of age (42). Furthermore, this observation of an reduced striatal MSN cell numbers and reduced 17b-estradiol initial increase of D1 density corresponds to the initial levels described here, reduced 17b-estradiol levels may also increase of cell size of the MSNs followed by atrophy at the contribute to the lower home-cage activity scores observed symptomatic stage of disease. Surprisingly, but analogous to in male tgHD rats. As IGFR1 was found to be differentially the lack of neuronal cell loss and cell-size alterations, we regulated in the gene-expression profiling data provided by were unable to detect a reduction of D1 receptor density in Hodges et al. (17), another level of regulation via female tgHD rats. 17b-estradiol signaling and interaction with IGFR1 (52) is As we found impaired motor function as well as striatal loss indicated. of MSNs only in males, our data strengthen the concept that Furthermore, a non-direct ER regulation might also occur sex hormones exert more potent neuroprotective effects in via the transcriptional co-activator PGC-1a as well as the female tgHD rats. This has already been shown in different co-chaperone BAG1. Although PGC-1a has been reported to neurological diseases, such as ischemic brain injury (43), play an important role in the transcriptional dysregulation traumatic neuronal damage (8) and Parkinson’s disease (44). and mitochondial dysfunction in HD (56–58) and, quite Furthermore, a protection from neurotoxicity by 17b-estradiol recently, BAG1 has been shown to ameliorate motor deficits has been reported in 3-nitropropionic acid and quinolinic- in tgHD mice in a sex-related manner (59), they were not acid-induced models of HD (10,45). Determination of periph- altered in array data from human HD striata (17). Interestingly, eral sex hormone levels revealed only a trend towards reduced analyses of those human data revealed that GABA-associated testosterone levels in tgHD males, whereas several other genes were differentially expressed between both sexes. studies have reported that the androgenic system was affected GABA receptors have been shown to interact with estrogens in HD (46–48). However, peripheral 17b-estradiol levels in (for review, see 50) and might thereby also account for the male tgHD rats were found to be significantly lower in com- observed increase of hyper-activity in tgHD female rats. parison to females as well as to wild-type littermates of the Potentially, therefore, these steroids could function not only same sex. It is known that steroid hormones released from as remote endocrine messengers, but might also act in a para- the adrenocortex and the gonads can pass the blood–brain crine manner to modify local neuronal activity by fine-tuning barrier and bind to intracellular receptors in the brain. GABAA receptor activity. Interestingly, the transcript However, the brain is also a steroidogenic organ. Certain SRD5A1 of the enzyme steroid-5-alpha-reductase, which con- neurons and glia in the central nervous system express verts progesterone into the active dihydroprogesterone enzymes (e.g. aromatase) that are required for the local syn- (5-alpha-DHPROG) was found to be downregulated thesis of pregnane neurosteroids, either de novo or from per- only in male HD patient striatum (17) (see also ipherally derived sources. The steroid hormones bind to their Supplementary Material, Table S4). Last but not least, steroids Human Molecular Genetics, 2008, Vol. 17, No. 17 2605 Downloaded from https://academic.oup.com/hmg/article-abstract/17/17/2595/632381 by guest on 27 June 2020

Figure 8. Summary of key findings. The onsets of phenotypic alterations in female tgHD rats are given above the timeline, while findings in the tgHD males are provided beneath. and neurosteroids determine gender-specific development of recommendations (62). All rats were kept under a 12:12 h the brain (60,61), an early process where altered steroid and light:dark cycle with lights on at 6 a.m., food (Altromin lab neurosteroid signaling might differentially affect the pheno- chow pellets, Altromin standard diet:1320; Lage, Germany) type of HD gene carriers. and tap water available ad libitum. Taken together, this is the first report describing a number All research and animal care procedures had been approved of sex differences in the rat model of HD. Together, these by the district government, Hannover, Germany, and followed differences illustrate a less severe course of the disease in principles described in the European Community’s Council female tgHD rats at the behavioral, physiological and neuro- Directive of 24 November 1986 (86/609/EEC). pathological level. As the decreased 17b-estradiol levels correlate with a loss of striatal DARPP32þ MSNs and as the MSNs do express ERs, our findings suggest the existence of Experimental design potent 17b-estradiol-mediated neuroprotective effects in the There were two sets of animals (N ¼ 40) studied until the ages striatum (Fig. 8). of 4 (set 1) and 14 (set 2) months, and with groups consisting of both female and male homozygous (þ/þ; n ¼ 10) and wild-type (2/2; n ¼ 10) rats. As not all investigations MATERIALS AND METHODS could be performed with all animals, the number of rats included into the analyses of each test is given in the legends. Animals Tg HD rats carrying a truncated huntingtin cDNA fragment Body weight, feeding and drinking with 51 CAG repeats under control of the native rat huntingtin promoter and their wild-type littermates were used in this Body weight was measured once a month. Before and after study (13). The expressed gene product was 75 kDa, corre- each home-cage activity test, the consumption of food sponding to 22% of the full-length huntingtin (cDNA position pellets and drinking water was determined individually as 324–2321, corresponding to exons 1–16), which are under well as absolutely and the relative food intake was calculated. the control of 886 bp of the rat huntingtin promoter (position Afterwards, the mean consumption (absolute food intake/per 900 to 15). day) and the percentage food intake (food intake/100-g body A colony of tg Sprague–Dawley rats was established at the weight) of each rat was calculated. central animal facilities, Hannover Medical School, and the Furthermore, indirect calorimetry was performed using a line was maintained by backcrossing. For experiments, LabMaster system (TSE-Systems, Bad Homburg, Germany) animals derived from generation F12 were used. at the age of 12 months. This system measures activity by After genotyping, rats were housed in gender- and means of light beam breaks for x-, y- and z-axis, drinking genotype-matched groups of two, according to FELASA and feeding online and oxygen consumption as well as 2606 Human Molecular Genetics, 2008, Vol. 17, No. 17 carbon dioxide production, allowing calculation of calori- Sacrifice and collection of tissues metric metabolism. The two sets of animals, at the age of 4 and 14 months, were deeply anesthetized by an i.p. injection of 0.5 ml liquid contain- ing 0.03 mg/kg body weight Medetomidin (Pfizer, Zurich, Behavioral testing Switzerland) and 15 mg/kg body weight Ketamin (Gra¨ub, Bern, Switzerland). The thoracic wall was opened and the left Based on previous work (13,25,63), sensitive and well- ventricle was cannulated. From the right ventricle, blood was validated behavioral assays were selected, following general drawn into tubes prefilled with ETDA and 500 000 kIU/l of principles of comprehensive phenotyping of rodents (64,65). Aprotinin (Celliance, Toronto, Canada). All samples were immediately cooled to 48C and centrifuged at 2000 rpm for

5 min (Eppendorf, Hamburg, Germany) in order to obtain Downloaded from https://academic.oup.com/hmg/article-abstract/17/17/2595/632381 by guest on 27 June 2020 Home-cage activity test pellet-free plasma. The supernatant was again centrifuged at 4000 rpm and subsequently at 8000 rpm, for 5 min each. To investigate the basic activity, the infrared-sensor-based Afterwards, all samples were stored at 2808C. home-cage activity test was carried out every month Under deep i.p. anesthesia, the animals were transcardially between the second and the 14th months, as previously perfused with 200 ml 4% paraformaldehyde (PFA), after a described (66). In brief, infrared sensors scored the number preflush of 15 ml phosphate-buffered saline (PBS). Brains and duration of small (,1 s) and large movements (.1 s). were removed and stored in 4% PFA tubes at 48C. After six Then, the mean activity per hour (in 15 min intervals) over hours, the brains were stored in tubes with 20% sucrose in 22 h was calculated for ‘duration of movements’ as well as PBS for further analysis. for ‘number of movements’.

Social interaction test of anxiety Determination of plasma sex hormone levels The social interaction test has been validated for testing anxiety of rats (67) and represents one of the few behavioral Plasma levels of 17b-estradiol, DHEA, testosterone and tests which are repeatedly applicable without losing construct estrone were determined by enzyme-linked immunosorbent validity (68). Two rats from different cages but with the same assays. The detection limits for these hormones were as gender and genotype were exposed to a novel environment in a follows: for 17b-estradiol, 4.6 pg/ml (IBL, Hamburg, sound isolation dimly lit box. The social interaction test was Germany); for DHEA, 0.10 ng/ml (Diagnostic Systems Lab- carried out every other month according to File and Seth oratory, Webster, Texas); for testosterone, 0.2 ng/ml (IBL); (67) with minor modifications. In brief, sociopositive beha- for estrone, 10 pg/ml (IBL). The interassay and intraassay viors within a 50 50 cm2 open-field arena located within coefficients of variation were below 10%. The numbers per an isolation box were monitored online using a video group vary from 8 to 10 animals. camera over a period of 10 min. Sums of sociopositive beha- viors such as sniffing/licking, grooming, crawling as well as following in the novel environment were used as a marker for a state of anxiety. Increased time spent in active social Immunohistochemistry interaction was interpreted as anxiolytic-like behavior (25,66). Right hemispheres were coronally cut into eight series of 40 mm thick sections using a freezing microtome. Free-floating sections were washed with 0.1 M PBS and then incubated for Accelerod test 15 min in a solution with 3% H2O2 and 10% methanol in The accelerod has been shown to be sensitive in detecting 0.1 M PBS. Tissue was then pre-blocked in 0.1 M PBS contain- functional deficits in motor coordination and balance (69) ing 0.3% Triton X-100 and 10% goat serum for 1 h. Sections and to produce more consistent results compared with the were subsequently incubated overnight at room temperature ‘classical’ rotarod (70). Furthermore, previous work revealed with a primary antibody (rabbit IgG polyclonal the test as sensitive in detecting motor deficits of tgHD rats anti-DARPP-32, 1:1000, AB1656, Chemicon, Billerica, MA, (13). To determine fore- and hind limb motor coordination USA) in 0.1 M PBS containing 0.3% Triton X-100 and 2.5% and balance, we used an accelerod apparatus for rats (TSE- goat serum. The tissue was then washed and incubated with a Systems, Bad Homburg, Germany) set to accelerate the rod secondary antibody (biotinylated goat anti-rabbit IgG 1:200, from 4 to 40 rpm in a period of 5 min as previously described E0353, DAKO, Stockholm, Sweden) for 1 h. After rinsing, (25). In brief, after 2 days of training, each rat was tested twice the color reaction for light microscopy was performed using on the test day. The better of these two trials was used for first the ABC solution (VectaStain Elite Kit, PK-6100, Vector further analysis of time spent on the accelerating rod and the Laboratories, Ja¨rfa¨lla, Sweden) and then adding the DAB kit maximum speed levels achieved. In both the wild-type as (Peroxidase Subtrate Kit, SK-4100, Vector Laboratories) well as in the tgHD groups, one male had to be excluded according to the manufacturer’s instructions, resulting in a from analyses, because they did not show any motivation to brown staining. The tissue was then rinsed in PBS and perform the test. mounted on gelatine-coated slides. Human Molecular Genetics, 2008, Vol. 17, No. 17 2607

Stereology-based quantification of DARPP321 neurons using a Nikon light microscope (Eclipse 80i; Nikon, Tokyo, and calculation of striatal volumes Japan), with a dedicated Nikon HiSN fluorescence system and imaging software (Stereo Investigator, MicroBrightField, Quantitative analysis was performed with a stereology work- station (stereology software from Visiopharm, Integrater Williston, VT, USA). System, Horsholm, Denmark). Striatal volumes were investi- gated in the neostriatal area between 11.28 mm Interaural/ 2.28 mm Bregma and 8.28 mm Interaural/20.72 mm Re-analysis of human gene-expression profiling data Bregma according to Paxinos (71) and as described previously Gene expression profiling data ((17), NCBI Gene Expression (15). The total numbers of DARPP32þ neurons were assessed Omnibus Data GSE 3790) were split by sex and matched using stereolgical principles (13 sections per brain, thickness sets (n ¼ 7 per group) of male controls, female controls, 40 mm, fraction 0.75% and frame 10%) (15). Cross-sectional

male HD and female HD patients served for further analysis. Downloaded from https://academic.oup.com/hmg/article-abstract/17/17/2595/632381 by guest on 27 June 2020 cell body areas of DARPP32þ neurons were investigated Statistical analyses of gene expression measures for included using the nucleator principle (72) in all profiles included in chips were carried out with the ArrayAssist 5.1 software (Stra- the assessment of total number of neurons (around 300 tagene). Gene expression was quantified by GC-RMA normal- neurons/rat). According to Ouimet et al. (34), more than ization with the ArrayAssist software package. To identify 95% of all DARPP32þ neurons are MSNs, which corresponds genes differentially expressed between HD and controls as with our findings. well as between female and male controls/patients, we used a t-statistics with Benjamini–Hochberg multiple testing cor- rection. The following data were included into the analysis: Autoradiography Significant differences of pair-wise comparisons on Whole brains were rapidly removed, blotted free of excess ‘hypothesis-driven’ genes using the ArrayAssist software blood and immediately frozen in 2-methylbutane (2508C). gene function tool and focusing on transcripts associated Subsequently, brains were cut on a cryostat microtome (CM with the topics ‘nuclear receptors’ and ‘estrogen related tran- 3050, Leica, Wetzlar, Germany; section thickness: 20 mm) scripts’ (yielding 164 transcripts with, e.g. ERs and CYP19 at 2208C. Rat brain slices were stained for dopamine D1 being among them; Supplementary Material, Tables S1–S4) and D2 receptors as previously described (42). For analyses, and ‘non-hypothesis-driven’ genes using a 1.5-fold change sections were placed on phosphor imaging plates (BAS-TR and P , 0.05 as criteria (Supplementary Material, Tables 2025, Raytest-Fuji, Straubenhardt, Germany) along with S5–S8). industrial tritium activity standards (Microscales; Amersham Biosciences, Freiburg, Germany). Upon exposure, the imaging plates were scanned with a high-performance Statistical analysis imaging plate reader (BAS5000 BioImage Analyser, Raytest- Data were subjected to one- or two-way ANOVA with one Fuji), which provides a spatial resolution of 50 mm. The evalua- between-subject factor (genotype) and with repeated measure- tion of digital receptor autoradiography was processed ments on one or more factors depending on the test used. The according to standard image analysis software (AIDA 2.31, Fisher PLSD test was used for post hoc comparison. Raytest-Fuji). Regressions were calculated using linear, simple regression analyses. A critical value for significance of P , 0.05 was used throughout the study. All data represent means + SEM. Immunofluorescence Brains were coronally cut into 30 mm thick sections in eight series using a freezing microtome. Free floating sections SUPPLEMENTARY MATERIAL were blocked with 10% bovine serum albumin (BSA) for 60 min, and afterwards washed twice with PBS for 10 min. Supplementary material is available at HMG online. Thereafter, sections were incubated overnight at 48C with the primary antibody against DARPP32 (rabbit monoclonal, 1:1000, EP720Y, Epitomics, Burlingame, Ca, USA) in combi- ACKNOWLEDGEMENTS nation with either anti-ER-a (mouse IgG1 monoclonal, 1:250, The authors thank Susanne Kuhlmann, Susanne Faßbender, ab858, Biozol, Eching, Germany) or anti-ER-b (mouse IgM Magdalene Vo¨geling and Jenny Stiller for their excellent tech- monoclonal, 1:250, Ab-2 9.88, Calbiochem, Gibbstown, NJ, nical assistance. We gratefully acknowledge the English cor- USA) in 1 M PBS containing 2% BSA and 0.3% Triton rections by Sheila Fryk. X-100. After two washing steps, sections were incubated with secondary antibodies (against DARPP32: Cy3 goat anti- Conflict of Interest statement. None declared. rabbit IgG, 1:1200, Jackson ImmunoResearch, Suffolk, UK; against ER-a: Alexa 488 donkey anti-mouse IgG, 1:200, Invi- trogen, Karlsruhe, Germany; against ER-b: Alexa 488 goat FUNDING anti-mouse IgM, 1:200, Invitrogen) for 2 h at room tempera- ture. After washing again, nuclei were stained HighQ-Foundation (to S.v.H); European Community ‘RAT- [4’,6-diamidino-2-phenylindole (DAPI), 1:1000, Invitrogen], stream STREP project’ (037846 to O.R. and S.v.H.); washed, mounted and cover-slipped. Sections were analyzed Swedish Research Council (M2006-6238 to A˚ .P.). 2608 Human Molecular Genetics, 2008, Vol. 17, No. 17

REFERENCES (2003) Immunolocalization of estrogen receptor beta in the mouse brain: comparison with estrogen receptor alpha. Endocrinology, 144, 1. The Huntington’s Disease Collaborative Research Group (1993) A novel 2055–2067. gene containing a trinucleotide repeat that is expanded and unstable on 21. Singh, M., Dykens, J.A. and Simpkins, J.W. (2006) Novel mechanisms Huntington’s disease . The Huntington’s Disease for estrogen-induced neuroprotection. Exp. Biol. Med. (Maywood), 231, Collaborative Research Group. Cell, 72, 971–983. 514–521. 2. Vonsattel, J.P. and DiFiglia, M. (1998) Huntington disease. 22. Mochel, F., Charles, P., Seguin, F., Barritault, J., Coussieu, C., Perin, L., J. Neuropathol. Exp. Neurol., 57, 369–384. Le Bouc, Y., Gervais, C., Carcelain, G., Vassault, A. et al. (2007) Early 3. Sarfarazi, M., Quarrell, O.W., Wolak, G. and Harper, P.S. (1987) An energy deficit in Huntington disease: identification of a plasma biomarker integrated microcomputer system to maintain a genetic register for traceable during disease progression. PLoS ONE, 2, e647. Huntington disease. Am. J. Med. Genet., 28, 999–1006. 23. van der Burg, J.M., Bacos, K., Wood, N.I., Lindqvist, A., Wierup, N., 4. Pearson, J.S., Petersen, M.C., Lazarte, J.A., Blodgett, H.E. and Kley, I.B. Woodman, B., Wamsteeker, J.I., Smith, R., Deierborg, T., Kuhar, M.J. (1955) An educational approach to the social problem of Huntington’s et al. (2008) Increased metabolism in the R6/2 mouse model of chorea. Proc. Staff Meet Mayo Clin., 30, 349–357.

Huntington’s disease. Neurobiol. Dis., 29, 41–51. Downloaded from https://academic.oup.com/hmg/article-abstract/17/17/2595/632381 by guest on 27 June 2020 5. Roos, R.A., Vegter-van der Vlis, M., Hermans, J., Elshove, H.M., Moll, 24. Bates, G., Harper, P. and Jones, L. (2002) Huntington’s Disease, 3rd edn. A.C., van de Kamp, J.J. and Bruyn, G.W. (1991) Age at onset in Oxford Monographs on Medical Genetics 45, Oxford University Press. Huntington’s disease: effect of line of inheritance and patient’s sex. 25. Nguyen, H.P., Kobbe, P., Rahne, H., Worpel, T., Jager, B., Stephan, M., J. Med. Genet., 28, 515–519. Pabst, R., Holzmann, C., Riess, O., Korr, H. et al. (2006) Behavioral 6. Foroud, T., Gray, J., Ivashina, J. and Conneally, P.M. (1999) Differences abnormalities precede neuropathological markers in rats transgenic for in duration of Huntington’s disease based on age at onset. J. Neurol. Huntington’s disease. Hum. Mol. Genet., 15, 3177–3194. Neurosurg. Psychiatry, 66, 52–56. 26. File, S.E., Mahal, A., Mangiarini, L. and Bates, G.P. (1998) Striking 7. Czlonkowska, A., Ciesielska, A., Gromadzka, G. and changes in anxiety in Huntington’s disease transgenic mice. Brain Res., Kurkowska-Jastrzebska, I. (2006) Gender differences in neurological 805, 234–240. disease: role of estrogens and cytokines. Endocrine, 29, 243–256. 27. Petrasch-Parwez, E., Nguyen, H.P., Lobbecke-Schumacher, M., Habbes, 8. Hoffman, G.E., Merchenthaler, I. and Zup, S.L. (2006) Neuroprotection H.W., Wieczorek, S., Riess, O., Andres, K.H., Dermietzel, R. and Von by ovarian hormones in animal models of neurological disease. Horsten, S. (2007) Cellular and subcellular localization of Huntington Endocrine, 29, 217–231. aggregates in the brain of a rat transgenic for Huntington disease. 9. Dhandapani, K.M. and Brann, D.W. (2002) Protective effects of estrogen J. Comp. Neurol., 501, 716–730. and selective estrogen receptor modulators in the brain. Biol. Reprod., 67, 28. Bjorkqvist, M., Petersen, A., Bacos, K., Isaacs, J., Norlen, P., Gil, J., 1379–1385. Popovic, N., Sundler, F., Bates, G.P., Tabrizi, S.J. et al. (2006) 10. Tunez, I., Collado, J.A., Medina, F.J., Pena, J., Del, C.M.M., Jimena, I., Progressive alterations in the hypothalamic–pituitary–adrenal axis in the Franco, F., Rueda, I., Feijoo, M., Muntane, J. et al. (2006) R6/2 transgenic mouse model of Huntington’s disease. Hum. Mol. Genet., 17beta-Estradiol may affect vulnerability of striatum in a 3-nitropropionic 15, 1713–1721. acid-induced experimental model of Huntington’s disease in 29. Petersen, A. and Bjorkqvist, M. (2006) Hypothalamic–endocrine aspects ovariectomized rats. Neurochem. Int., 48, 367–373. in Huntington’s disease. Eur. J. Neurosci., 24, 961–967. 11. Stroppolo, A., Tian, C., Guinea, B., Olm, V., Sheffield, R., Sommer, J. and 30. Giordano, M. and Mejia-Viggiano, M.C. (2001) Gender differences in Ehrlich, M.E. (2004) 17beta-Estradiol promotes striatal medium size spontaneous and MK-801-induced activity after striatal lesions. Brain Res. spiny neuronal maturation in vitro. Neuroendocrinology, 79, 259–267. Bull., 56, 553–561. 12. Holzmann, C., Maueler, W., Petersohn, D., Schmidt, T., Thiel, G., Epplen, 31. Dorner, J.L., Miller, B.R., Barton, S.J., Brock, T.J. and Rebec, G.V. J.T. and Riess, O. (1998) Isolation and characterization of the rat (2007) Sex differences in behavior and striatal ascorbate release in the 140 huntingtin promoter. Biochem. J., 336, 227–234. CAG knock-in mouse model of Huntington’s disease. Behav. Brain Res., 13. von Horsten, S., Schmitt, I., Nguyen, H.P., Holzmann, C., Schmidt, T., 178, 90–97. Walther, T., Bader, M., Pabst, R., Kobbe, P., Krotova, J. et al. (2003) 32. Slow, E.J., van Raamsdonk, J., Rogers, D., Coleman, S.H., Graham, R.K., Transgenic rat model of Huntington’s disease. Hum. Mol. Genet., 12, Deng, Y., Oh, R., Bissada, N., Hossain, S.M., Yang, Y.Z. et al. (2003) 617–624. Selective striatal neuronal loss in a YAC128 mouse model of Huntington 14. Temel, Y., Cao, C., Vlamings, R., Blokland, A., Ozen, H., Steinbusch, disease. Hum. Mol. Genet., 12, 1555–1567. H.W., Michelsen, K.A., von Horsten, S., Schmitz, C. and 33. Carter, R.J., Lione, L.A., Humby, T., Mangiarini, L., Mahal, A., Bates, Visser-Vandewalle, V. (2006) Motor and cognitive improvement by deep G.P., Dunnett, S.B. and Morton, A.J. (1999) Characterization of brain stimulation in a transgenic rat model of Huntington’s disease. progressive motor deficits in mice transgenic for the human Huntington’s Neurosci. Lett., 406, 138–141. disease mutation. J. Neurosci., 19, 3248–3257. 15. Kantor, O., Temel, Y., Holzmann, C., Raber, K., Nguyen, H.P., Cao, C., 34. Ouimet, C.C., Langley-Gullion, K.C. and Greengard, P. (1998) Turkoglu, H.O., Rutten, B.P., Visser-Vandewalle, V., Steinbusch, H.W. Quantitative immunocytochemistry of DARPP-32-expressing neurons in et al. (2006) Selective striatal neuron loss and alterations in behavior the rat caudatoputamen. Brain Res., 808, 8–12. correlate with impaired striatal function in Huntington’s disease 35. Turmaine, M., Raza, A., Mahal, A., Mangiarini, L., Bates, G.P. and transgenic rats. Neurobiol. Dis., 22, 538–547. Davies, S.W. (2000) Nonapoptotic neurodegeneration in a transgenic 16. Cao, C., Temel, Y., Blokland, A., Ozen, H., Steinbusch, H.W., Vlamings, mouse model of Huntington’s disease. Proc. Natl. Acad. Sci. USA, 97, R., Nguyen, H.P., von Horsten, S., Schmitz, C. and Visser-Vandewalle, V. 8093–8097. (2006) Progressive deterioration of reaction time performance and 36. Li, X.J. (1999) The early cellular pathology of Huntington’s disease. Mol. choreiform symptoms in a new Huntington’s disease transgenic ratmodel. Neurobiol., 20, 111–124. Behav. Brain Res., 170, 257–261. 37. Nopoulos, P., Magnotta, V.A., Mikos, A., Paulson, H., Andreasen, N.C. 17. Hodges, A., Strand, A.D., Aragaki, A.K., Kuhn, A., Sengstag, T., Hughes, and Paulsen, J.S. (2007) Morphology of the cerebral cortex in preclinical G., Elliston, L.A., Hartog, C., Goldstein, D.R., Thu, D. et al. (2006) Huntington’s disease. Am. J. Psychiatry, 164, 1428–1434. Regional and cellular gene expression changes in human Huntington’s 38. Svenningsson, P., Nishi, A., Fisone, G., Girault, J.A., Nairn, A.C. and disease brain. Hum. Mol. Genet., 15, 965–977. Greengard, P. (2004) DARPP-32: an integrator of neurotransmission. 18. Langley, K.C., Bergson, C., Greengard, P. and Ouimet, C.C. (1997) Annu. Rev. Pharmacol. Toxicol., 44, 269–296. Co-localization of the D1 dopamine receptor in a subset of 39. Auger, A.P., Meredith, J.M., Snyder, G.L. and Blaustein, J.D. (2001) DARPP-32-containing neurons in rat caudate-putamen. Neuroscience, 78, Oestradiol increases phosphorylation of a dopamine- and cyclic 977–983. AMP-regulated phosphoprotein (DARPP-32) in female rat brain. 19. Augood, S.J., Faull, R.L. and Emson, P.C. (1997) Dopamine D1 and D2 J. Neuroendocrinol., 13, 761–768. receptor gene expression in the striatum in Huntington’s disease. Ann. 40. Lindskog, M., Svenningsson, P., Fredholm, B.B., Greengard, P. and Neurol., 42, 215–221. Fisone, G. (1999) Activation of dopamine D2 receptors decreases 20. Mitra, S.W., Hoskin, E., Yudkovitz, J., Pear, L., Wilkinson, H.A., DARPP-32 phosphorylation in striatonigral and striatopallidal projection Hayashi, S., Pfaff, D.W., Ogawa, S., Rohrer, S.P., Schaeffer, J.M. et al. neurons via different mechanisms. Neuroscience, 88, 1005–1008. Human Molecular Genetics, 2008, Vol. 17, No. 17 2609

41. Nishi, A., Snyder, G.L. and Greengard, P. (1997) Bidirectional regulation of 57. Tcherepanova, I., Puigserver, P., Norris, J.D., Spiegelman, B.M. and DARPP-32 phosphorylation by dopamine. J. Neurosci., 17, 8147–8155. McDonnell, D.P. (2000) Modulation of estrogen receptor-alpha 42. Bauer, A., Zilles, K., Matusch, A., Holzmann, C., Riess, O. and von transcriptional activity by the coactivator PGC-1. J. Biol. Chem., 275, Horsten, S. (2005) Regional and subtype selective changes of 16302–16308. neurotransmitter receptor density in a rat transgenic for the Huntington’s 58. Weydt, P., Pineda, V.V., Torrence, A.E., Libby, R.T., Satterfield, T.F., disease mutation. J. Neurochem., 94, 639–650. Lazarowski, E.R., Gilbert, M.L., Morton, G.J., Bammler, T.K., Strand, 43. Dubal, D.B., Rau, S.W., Shughrue, P.J., Zhu, H., Yu, J., Cashion, A.B., A.D. et al. (2006) Thermoregulatory and metabolic defects in Suzuki, S., Gerhold, L.M., Bottner, M.B., Dubal, S.B. et al. (2006) Huntington’s disease transgenic mice implicate PGC-1alpha in Differential modulation of estrogen receptors (ERs) in ischemic brain Huntington’s disease neurodegeneration. Cell Metab., 4, 349–362. injury: a role for ERalpha in estradiol-mediated protection against delayed 59. Orr, A.L., Huang, S., Roberts, M.A., Reed, J.C., Li, S. and Li, X.J. (2008) cell death. Endocrinology, 147, 3076–3084. Sex-dependent effect of bag1 in ameliorating motor deficits of 44. Gillies, G.E., Murray, H.E., Dexter, D. and McArthur, S. (2004) Sex Huntington’s disease transgenic mice. J. Biol. Chem, 283, 16027–16036. dimorphisms in the neuroprotective effects of estrogen in an animal model 60. McCarthy, M.M. (2008) Estradiol and the developing brain. Physiol. Rev., 88, 91–124. of Parkinson’s disease. Pharmacol. Biochem. Behav., 78, 513–522. Downloaded from https://academic.oup.com/hmg/article-abstract/17/17/2595/632381 by guest on 27 June 2020 45. Heron, P. and Daya, S. (2000) 17Beta-estradiol protects against quinolinic 61. McCarthy, M.M., Auger, A.P. and Perrot-Sinal, T.S. (2002) Getting acid-induced lipid peroxidation in the rat brain. Metab. Brain Dis., 15, excited about GABA and sex differences in the brain. Trends Neurosci., 267–274. 25, 307–312. 46. Van Raamsdonk, J.M., Murphy, Z., Selva, D.M., Hamidizadeh, R., 62. Rehbinder, C., Baneux, P., Forbes, D., van Herck, H., Nicklas, W., Pearson, J., Petersen, A., Bjorkqvist, M., Muir, C., Mackenzie, I.R., Rugaya, Z. and Winkler, G. (1996) FELASA recommendations for the Hammond, G.L. et al. (2007) Testicular degeneration in Huntington health monitoring of mouse, rat, hamster, gerbil, guinea pig and rabbit disease. Neurobiol. Dis., 26, 512–520. experimental units. Report of the Federation of European Laboratory 47. Papalexi, E., Persson, A., Bjorkqvist, M., Petersen, A., Woodman, B., Animal Science Associations (FELASA) Working Group on Animal Bates, G.P., Sundler, F., Mulder, H., Brundin, P. and Popovic, N. (2005) Health accepted by the FELASA Board of Management, November 1995. Reduction of GnRH and infertility in the R6/2 mouse model of Lab. Anim., 30, 193–208. 63. Karl, T., Hoffmann, T., Pabst, R. and von Horsten, S. (2003) Extreme Huntington’s disease. Eur. J. Neurosci., 22, 1541–1546. reduction of dipeptidyl peptidase IV activity in F344 rat substrains is 48. Markianos, M., Panas, M., Kalfakis, N. and Vassilopoulos, D. (2005) associated with various behavioral differences. Physiol. Behav., 80, Plasma testosterone in male patients with Huntington’s disease: relations 123–134. to severity of illness and dementia. Ann. Neurol., 57, 520–525. 64. Crawley, J.N. (1999) Behavioral phenotyping of transgenic and knockout 49. Belelli, D. and Lambert, J.J. (2005) Neurosteroids: endogenous regulators mice: experimental design and evaluation of general health, sensory of the GABA(A) receptor. Nat. Rev. Neurosci., 6, 565–575. functions, motor abilities, and specific behavioral tests. Brain Res., 835, 50. Joels, M. (1997) Steroid hormones and excitability in the mammalian 18–26. brain. Front. Neuroendocrinol., 18, 2–48. 65. Karl, T., Pabst, R. and von Horsten, S. (2003) Behavioral phenotyping of 51. Bogush, A., Pedrini, S., Pelta-Heller, J., Chan, T., Yang, Q., Mao, Z., mice in pharmacological and toxicological research. Exp. Toxicol. Sluzas, E., Gieringer, T. and Ehrlich, M.E. (2007) AKT and CDK5/p35 Pathol., 55, 69–83. mediate brain-derived neurotrophic factor induction of DARPP-32 in 66. Breivik, T., Stephan, M., Brabant, G.E., Straub, R.H., Pabst, R. and von medium size spiny neurons in vitro. J. Biol. Chem., 282, 7352–7359. Horsten, S. (2002) Postnatal lipopolysaccharide-induced illness 52. Cardona-Gomez, G.P., Mendez, P., DonCarlos, L.L., Azcoitia, I. and predisposes to periodontal disease in adulthood. Brain Behav. Immun., 16, Garcia-Segura, L.M. (2001) Interactions of estrogens and insulin-like 421–438. growth factor-I in the brain: implications for neuroprotection. Brain Res. 67. File, S.E. and Seth, P. (2003) A review of 25 years of the social interaction Brain Res. Rev., 37, 320–334. test. Eur. J. Pharmacol., 463, 35–53. 53. Rune, G.M. and Frotscher, M. (2005) Neurosteroid synthesis in the 68. File, S.E. and Hyde, J.R. (1978) Can social interaction be used to measure hippocampus: role in synaptic plasticity. Neuroscience, 136, 833–842. anxiety? Br. J. Pharmacol., 62, 19–24. 54. McCullough, L.D., Blizzard, K., Simpson, E.R., Oz, O.K. and Hurn, P.D. 69. Bogo, V., Hill, T.A. and Young, R.W. (1981) Comparison of accelerod (2003) Aromatase cytochrome P450 and extragonadal estrogen play a role and rotarod sensitivity in detecting ethanol- and acrylamide-induced in ischemic neuroprotection. J. Neurosci., 23, 8701–8705. performance decrement in rats: review of experimental considerations of 55. Zwain, I.H. and Yen, S.S. (1999) Neurosteroidogenesis in astrocytes, rotating rod systems. Neurotoxicology, 2, 765–787. oligodendrocytes, and neurons of cerebral cortex of rat brain. 70. Jones, B.J. and Roberts, D.J. (1968) The quantiative measurement of Endocrinology, 140, 3843–3852. motor inco-ordination in naive mice using an acelerating rotarod. 56. St-Pierre, J., Drori, S., Uldry, M., Silvaggi, J.M., Rhee, J., Jager, S., J. Pharm. Pharmacol., 20, 302–304. Handschin, C., Zheng, K., Lin, J., Yang, W. et al. (2006) Suppression of 71. Paxinos, G. and Watson, C. (1998) The Rat Brain in Stereotaxic reactive oxygen species and neurodegeneration by the PGC-1 Coordinates. San Diego: Academic Press. transcriptional coactivators. Cell, 127, 397–408. 72. Gundersen, H.J. (1988) The nucleator. J. Microsc., 151, 3–21.